INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Reactive oxygen species (ROS) are a group of very reactive, short-lived chemicals produced during normal respiratory processes or after oxidative insults. ROS include superoxide anion, hydrogen peroxide, hydroxyl radical, and organic peroxides, among others (56). ROS, at low concentrations, have been implicated in the regulation of several physiological processes such as proliferation (55), differentiation (3), apoptosis (25), and senescence (14). At high concentrations, ROS are highly toxic to cells, by inducing DNA damage, lipid peroxidation, and protein degradation (56). Several lines of evidence suggest that ROS may mediate apoptosis: (i) the production of ROS or depletion of antioxidants promotes apoptosis (43, 64); (ii) antioxidants inhibit apoptosis (43, 64), and (iii) generation of ROS mediates p53-induced apoptosis (48). ROS appear to function mainly at the initiation/activation step of apoptosis since most apoptosis inducers produce ROS. Examples include irradiation, UV, chemicals, ceramides, and growth factor withdrawal, among many others (28, 66). How do ROS, as the highly reactive but nonspecific molecules, mediate well-coordinated apoptosis? It is unlikely that ROS themselves are direct signaling molecules that activate some crucial components of apoptosis machinery due to their lack of biological specificity. More likely, ROS act indirectly by modifying cellular redox-sensitive molecules, such as p53 and NF-B, that are directly involved in apoptosis (6, 42, 58, 61).

Metal chelators and metal ions are redox-sensitive agents that could mediate ROS generation through the Fenton reaction. 1,10-Phenanthroline (OP), a typical metal chelator (4), has been previously shown to either induce or suppress apoptosis in a cell line-dependent manner (1, 2, 7, 41). OP induces apoptosis by increasing expression of cell surface APO/Fas ligand (41), by chelating zinc (2, 26), or by chelating copper and promoting its redox activity to induce internucleosomal DNA fragmentation (11, 67). Copper ion (Cu²⁺) is a highly reactive ion that has been used in concert with OP, chemicals, or carcinogens to induce apoptosis in a number of cellular models (11, 24, 51, 53, 67, 70). Cu²⁺ is readily reduced to Cu⁺, and the latter reacts with H₂O₂ through the Fenton reaction to form the highly toxic hydroxyl radical that causes cell death (71). The zinc ion, however, is a rather stable trace element. It has been found that zinc can either induce or suppress apoptosis in a concentration-dependent manner (19, 49). High concentrations of extracellular zinc (500 to 1,000 µM) suppress apoptosis, probably by either inhibiting a Ca²⁺/Mg²⁺-dependent endonuclease, which is responsible for DNA fragmentation (73), or by inhibiting caspase 3 (47). At lower concentrations (80 to 200 µM), zinc appears to induce apoptosis (19, 49), most likely by enhancing the generation of hydroxyl free radicals (46). Furthermore, zinc induces neuronal death both in cultured cells (12, 31, 39) and in animals (32).

To counteract the damaging effects of ROS, aerobic cells are endowed with extensive antioxidant defense systems. These defense systems consist mainly of antioxidant enzymes (e.g., superoxide dismutase, catalase, glutathione peroxidase, and glutathione reductase), antioxidant proteins (e.g., thioredoxin and metallothionein [MT]), and small molecular antioxidants (e.g., glutathione, N-acetyl-L-cysteine, and vitamin C). Almost all antioxidant molecules have been shown to protect cells against apoptosis induced by redox-sensitive reagents (5, 29, 33, 40).

In an attempt to understand signaling pathways leading to apoptosis induced by OP, a metal-chelating and redox-sensitive reagent (58), we used the differential display (DD) technique for identification of OP-responsive genes and reported the cloning of an OP-inducible gene that encodes glutathione synthetase (GSS) (57). Here we report the cloning and characterization of second OP-inducible gene which encodes SAG (sensitive to apoptosis gene), an evolutionarily conserved novel zinc RING finger protein with features of metal ion binding and free radical scavenging. When overexpressed in several human cell lines, SAG protects cells against apoptosis induced by OP, zinc, and copper. Thus, SAG appears to function as an antioxidant molecule to inhibit metal ion- or ROS-induced apoptosis.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Cell maintenance and drug treatment. Human neuroblastoma line SY5Y and human kidney line 293 were cultured in Dulbecco's modified Eagle medium containing 10% fetal calf serum (FCS; Sigma). Human colon carcinoma line DLD-1 was grown in Eagle minimal essential medium (MEM) containing 10% FCS. Two mouse tumor lines, L-RT101 and H-Tx, were used. L-RT101, a tumor promoter-transformed JB6 epidermal line (62), was cultured in Eagle MEM containing, FCS; H-Tx, a spontaneously transformed liver line (59a), was grown in Dulbecco's modified Eagle medium supplemented with 10% FCS and 1 mM sodium pyruvate. For drug treatment, subconfluent cells were exposed to dimethyl sulfoxide vehicle control, OP or metal ions, zinc sulfate, or copper sulfate (Sigma) for various periods of time up to 24 h.

DD. Analysis was performed by using an RNAimage kit B (GeneHunter) as instructed by the manufacturer, with slight modification (59). Briefly, total RNA was isolated by using RNAzol solution (Tel-Test) and subjected to reverse transcription followed by PCR. PCR fragments were resolved in sequencing gels. The fragments reproducibly showing differential expression were PCR amplified and used as probes for Northern analysis. Northern analysis-confirmed fragments were then subcloned into TA cloning vector (Invitrogen) and sequenced by using DNA Sequenase version 2.0 (Amersham). The GenBank search was performed with the Genetics Computer Group program.

cDNA library screening and 5' rapid amplification of cDNA ends. The mouse lung cDNA library (Stratagene) was screened with a DD fragment to clone full-length mouse SAG (mSAG). The longest clone isolated was a 1.0-kb fragment consisting of partial open reading frame and the entire 3'-end untranslated region. A mouse brain Marathon-Ready cDNA (Clontech) was used to clone the 5'-end upstream sequence, which yielded a 100-bp fragment consisting of the 5'-end untranslated sequence and some of the coding sequence. To clone human SAG (hSAG), the 1.0-kb mSAG fragment was used as a probe to screen a human HeLa cDNA library (Stratagene). A 0.75-kb fragment flanking the entire open reading frame and some 3'-end untranslated region was obtained.

Cellular localization by immunofluorescence. Cells were plated on coverslips in 24-well culture dishes and transfected by the calcium phosphate method with the following constructs: pcDNA3.1 (vector control with a Myc-His tag; Invitrogen); pcDNA3.1-SAG (SAG cDNA cloned upstream and in frame with Myc-His tag); or pcDNA3.1-LacZ. Two days posttransfection, cells were washed with cold phosphate-buffered saline (PBS) and fixed with 3% formaldehyde in PBS for 10 min followed by 5 min in methanol-acetone (1:1). The fixed cells were washed with PBS four times and incubated with Myc tag antibody (1:200 dilution; Invitrogen) in PBS containing 1% bovine serum albumin, 0.1% saponin, and 2 µg of 4',6-diamidino-2-phenylindole (DAPI) per ml for 1 h in the dark with shaking. Cells were then washed with 0.1% saponin in PBS and incubated with fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse antibody (1:100 dilution; Jackson Laboratory) for 1 h in the same condition as the first antibody. After incubation, cells were washed first with 0.1% saponin in PBS and then with PBS. The coverslips were mounted to glass slides with nonfade mounting medium and analyzed in a Leica Dialux 20 microscope.

Fluorescence in situ hybridization (FISH) chromosome mapping. (i) Slide preparation. Lymphocytes isolated from human blood were cultured in MEM supplemented with 10% FCS and phytohemagglutinin (PHA) at 37°C for 68 to 72 h. The lymphocyte cultures were treated with bromodeoxyuridine (0.18 mg/ml; Sigma) to synchronize the cell population. The synchronized cells were washed with serum-free medium and recultured at 37°C for 6 h in -MEM with thymidine (2.5 µg/ml; Sigma). Cell were harvested, and slides were made by using standard procedures including hypotonic treatment, fixation, and air drying.

(ii) In situ hybridization and FISH detection. A 750-bp hSAG cDNA probe was biotinylated with dATP, using the Bethesda Research Laboratories BioNick labeling kit (15°C, 1 h) (22). FISH detection was performed as described previously (22, 23). Briefly, slides were baked at 55°C for 1 h. After RNase treatment, the slides were denatured in 70% formamide in 2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate) for 2 min at 70°C followed by dehydration with ethanol. Probes were denatured at 75°C for 5 min in a hybridization mix consisting of 50% formamide and 10% dextran sulfate. Probes were loaded on the denatured chromosomal slides. After overnight hybridization, slides were washed and signals were detected. FISH signals and the DAPI banding pattern were recorded separately by taking photographs, and FISH mapping data were correlated with chromosomal bands by superimposing FISH signals with DAPI-banded chromosomes.

ESI-MS. Electrospray ionization mass spectrometry (ESI-MS) was performed with a double-focusing hybrid mass spectrometer (MAT 900Q; Finnigan, Bremen, Germany) with a mass-to-charge range of 10,000 at 5-kV full acceleration potential. A position-and-time-resolved-ion-counting (PATRIC) scanning array detector was used. An ESI interface based on a heated metal capillary inlet and a low-flow micro ESI source (analyte flow rate of 150 nl min¹) were used (50). The metal capillary temperature was maintained at around 150 to 200°C for metal-protein complex studies. Recombinant protein in a 7 M urea-denaturing solution was refolded by dialysis in 50 µM ZnSO₄ for 3 days with three changes of buffer. Prior to ESI-MS measurement, the SAG solution was washed with a solution of 10 mM ammonium bicarbonate (pH 7) and 1 mM dithiothreitol (DTT), and excess zinc was removed by centrifugal ultrafiltration by passage through a 10-kDa-molecular-mass cutoff centrifugal filtration cartridge (Microcon-10 microconcentrator; Amicon, Beverly, Mass.). For ESI-MS analysis, a small portion of the filtered SAG protein solution was diluted into either a denaturing solvent (80:15:5 [vol/vol/vol] acetonitrile-water-acetic acid [pH 2.5]) or a nondenaturing solution (10 mM ammonium bicarbonate, 1 mM DTT [pH 7]).

Assay for lipoprotein oxidation. Lipoproteins (100 µg of protein/ml; Intraocel) were incubated with 10 µM CuSO₄ or with 5 mM 2,2-azo-bis-2-amidinopropane hydrochloride (AAPH) for 4 h at 37°C in the presence of various concentrations of purified SAG protein. AAPH is a water-soluble azo compound that thermally decomposes and generates water-soluble peroxyl radicals at a constant rate (21). Oxidation was terminated by the addition of 10 µM butylated hydroxytoluene and refrigeration at 4°C. The extent of lipoprotein oxidation was measured by the thiobarbituric acid-reactive substances (TBARS) assay, using malondialdehyde for the standard curve, as described previously (10).

Antibody generation. Two polyclonal antibodies against hSAG protein were generated, using standard methods, by Zymed Laboratories, Inc. (San Francisco, Calif.). Briefly, the peptide antibody was generated as following. A 16-amino-acid peptide (SAG-Pepl; QNNRCPLCQQDWWQR) located in the C terminus of SAG protein (codons 95 to 110) was synthesized and purified via standard techniques. The purified peptide was conjugated to keyhole limpet hemocyanin via cysteine residues. The conjugated peptide (0.5 mg) was emulsified with an equal volume of complete Freund's adjuvant and subcutaneously injected into rabbits, followed by four boosts with 0.5 mg each in incomplete Freund's adjuvant at 3-week intervals. Rabbits were bled 10 days after the final boost, and antiserum was collected. The same protocol was used for protein antibody production using bacterially expressed and purified hSAG protein as the antigen.

Preparation of subcellular fractions. Subcellular fractions of the D1-6 and D12-1 cells or 293 transfectants were prepared as described previously (74). Briefly, the cells were harvested after exposure to ZnSO₄ or CuSO₄ for various periods of time and washed twice with ice-cold PBS. The cell pellet was resuspended in 5 volumes of cold buffer A (20 mM HEPES-KOH [pH 7.5], 10 mM KCl, 1.5 mM MgCl₂, 1 mM sodium EDTA, 1 mM sodium EGTA, 1 mM DTT, and 0.1 mM phenylmethylsulfonyl fluoride) containing 250 mM sucrose and freshly added protease inhibitor cocktail (Boehringer Mannheim). After homogenization for 10 strokes with a Dounce homogenizer, the nuclei were discarded by centrifugation at 1,000 × g for 10 min at 4°C. The supernatant was further centrifuged at 10,000 × g for 15 min at 4°C, and the resulting supernatants were used for protein and immunoblot assays.

Immunoblot assay. The cytosolic proteins (50 µg) prepared as described above were resolved on a sodium dodecyl sulfate-polyacrylamide gel and transferred to a polyvinylidene difluoride membrane. Immunoblots were probed with a monoclonal antibody against caspase 3 (Transduction Laboratories), a monoclonal antibody against human cytochrome c (PharMingen), or a rabbit polyclonal antibody against caspase 7 (15). For SAG detection, either rabbit anti-SAG antibody (1:1,000 dilution) or Flag antibody (Sigma) was used. Proteins were detected by horseradish peroxidase-conjugated secondary antibody coupled with enhanced chemiluminescence Western blotting detection reagents (Amersham).

Northern analysis and immunoprecipitation. The assays were performed as detailed previously (60, 62). Briefly, total RNA was isolated and subjected to Northern analysis using mSAG and hSAG cDNA as probes. For immunoprecipitation, subconfluent SAG transfectants and the vector control cells were subjected to methionine starvation for 1 h and metabolically labeled with Trans[³⁵S]-label (0.2 mCi/ml) for 3 h. Cells were then lysed on ice for 30 min in lysis buffer (60) and spun at 12,000 × g. The equal counts of trichloroacetic acid-precipitable radioactivity in the supernatant (3 × 10⁷ cpm) were immunoprecipitated with rabbit anti-human SAG antibody. The immunoprecipitates were then collected, washed, and analyzed by separation on a sodium dodecyl sulfate-10 to 20% polyacrylamide gel followed by autoradiography. Densitometric quantitation was performed on a densitometer (Molecular Dynamics).

Construction of SAG expression vectors and establishment of stable expressing clones. Reverse transcription-PCR (60) was performed to clone the hSAG coding region into the following expression vectors: pcDNA 3C (containing a Myc tag; Invitrogen), for immunolocalization study; and pcDNA3 (Invitrogen), for stable transfection in DLD-1 cells. To generate a Flag-tagged SAG expression construct, hSAG cDNA was PCR amplified by using primers Flag-SAG (5'-CG GGGTACCGCCATGGACTACAAGGACGACGATGACAAGGCCGACGT GGAAGAC-3') and SAG-XhoI (5'-CCGCTCGAGTCATTTGCCGATTCTTTGGACCAC-3'). The PCR fragment was subcloned into pcDNA3. All PCR-generated clones were verified by DNA sequencing for appropriate orientation and freedom of mutations. To establish stable SAG-expressing lines, the pcDNA3-SAG or pcDNA3-Flag-SAG construct, respectively, was transfected by Lipofectamine (Bethesda Research Laboratories), along with the vector control, into DLD-1 human colon carcinoma cells or SY5Y human neuroblastoma cells that express a low level of endogenous SAG. After G418 (600 µg/ml) selection, stable clones were ring isolated and SAG expression was monitored by Northern analysis; selected clones were then examined for protein expression by immunoprecipitation or Western analysis. For 293 cells, transient transfection was performed by the calcium phosphate method as described previously (58).

DNA fragmentation assay. Subconfluent cells (80 to 90%) were treated with OP (150 µM) or ZnSO₄ (125 µM) for 24 h. Both detached and attached cells were harvested by scraping with a rubber policeman. Cells were collected by centrifugation and lysed in lysis buffer (5 mM Tris-HCl [pH 8], 20 mM EDTA, 0.5% Triton X-100) on ice for 45 min. Fragmented DNA in the supernatant after centrifugation at 14,000 rpm (45 min at 4°C) was extracted twice with phenol-chloroform and once with chloroform and then precipitated with ethanol and salt. The DNA pellet was washed once with 70% ethanol and resuspended in Tris-EDTA buffer with 100 µg of RNase per ml at 37°C for 2 h. The fragmented DNA was separated in 1.8% agarose gel electrophoresis, stained with ethidium bromide, and visualized under UV light (58).

TUNEL assay. The terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) assay was performed as instructed by the manufacturer (Boehringer Mannheim). Briefly, 5 × 10⁴ cells were plated onto eight-well glass slides. After being treated with 1.25 mM copper (CuSO₄) or 200 µM zinc (ZnSO₄) for 16 h, cells were fixed with 0.5% glutaraldehyde for 10 min and then washed with PBS twice. The fixed cells were incubated in a permeabilization solution (0.1% Triton X-100, 0.1% sodium citrate) for 2 min on ice. The TUNEL reaction mixture (50 µl) was added to samples, which were incubated for 1 h at 37°C and then washed with PBS three times. Samples were embedded with antifade prior to analysis under a fluorescence microscope.

Nucleotide sequence accession numbers. The SAG sequences have been deposited in GenBank with accession no. AF092877 (mouse) and AF092878 (human). The cDNA clones have also been deposited in the American Type Culture Collection with assigned numbers 98402 (mouse) and 98405 (human).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Cloning of a novel protein that is evolutionarily conserved and contains a zinc RING finger domain. The DD technique was used in an attempt to isolate genes responsible for or associated with OP-induced apoptosis in two murine tumor lines (58). Two OP-inducible cDNA fragments were isolated. One encodes GSS, while the other is novel (57). The novel cDNA fragment was used as a probe to screen a mouse lung cDNA library. The resulting cDNA clone identified by this method contains a 1,140-bp insert encoding an open reading frame of 113 amino acids, including 12 cysteine residues (Fig. 1). The open reading frame was preceded by a 17-bp upstream sequence. An initiation codon was located in a context that conformed precisely 100% to the Kozak consensus sequence (34). The 3'-end untranslated region consists of 792 bp of sequence with two polyadenylation signals (AATAAA). Taken together, these features indicates that we have cloned a nearly full length cDNA. Since this clone was inducible in association with the OP-induced apoptosis pathway, we have named it SAG, for "sensitive to apoptosis gene."

View larger version (65K): [in this window] [in a new window]   FIG. 1.   SAG is evolutionarily conserved among different species. Primary amino acid sequences of human and mouse SAG were deduced from cDNAs cloned through DD and library screening. (A) Comparative alignments of SAG coding sequences from human, mouse, C. elegans, and yeast. Identity is shaded, and similarity is boxed. (B) Consensus sequence for the RING-H2 motif and comparison of the zinc RING finger domain of SAG with the RING-H2 motif. The C3H2C3 residues are in boldface.

SAG is inducible by OP in mouse tumor cells. To confirm that SAG is subject to OP induction, we performed Northern analysis with RNAs isolated from two mouse tumor lines, L-RT101 (tumor promoter-transformed JB6 epidermal cells)(62) and H-Tx (spontaneously transformed mouse liver cells) (59a). As shown in Fig. 2, mSAG cDNA detects an OP-inducible transcript of 1.2 kb in size. Induction occurs 1 h after drug treatment and persists for up to 24 h in both cell lines, indicating that SAG is an early-response gene.

View larger version (72K): [in this window] [in a new window]   FIG. 2.   SAG is inducible by OP. Subconfluent mouse L-RT101 and H-Tx cells were treated with OP (150 µM) for various periods of time up to 24 h and subjected to total RNA isolation and Northern analysis (with 15 µg of total RNA) using mouse SAG cDNA as a probe. Ethidium bromide staining of 28S and 18S rRNAs as loading controls is shown at the bottom.

Tissue distribution and cellular and chromosomal localization of SAG. hSAG expression in several human tissues was examined. As shown in Fig. 3, hSAG was ubiquitously expressed in all tissues examined, although relatively low levels were detected in brain, placenta, lung, and kidney. Very high mRNA levels were detected in the heart, skeletal muscle, and testis (lanes 1, 6, and 12).

View larger version (54K): [in this window] [in a new window]   FIG. 3.   SAG expression in multiple human tissues. hSAG cDNA was used as a probe for Northern analysis of poly(A)+ RNA isolated from different human tissues (Clontech). For an internal loading control, the housekeeping -actin gene was used (bottom).

View larger version (65K): [in this window] [in a new window]   FIG. 4.   Cellular and chromosomal localization of human SAG. (A to C) Cellular localization. NIH 3T3 cells were plated onto coverslips, transiently transfected with plasmid encoding SAG-Myc tag, and immunofluoresced by antibody against Myc tag as detailed in Materials and Methods. Shown are the SAG-expressing cells (A), vector control cells (B), and LacZ-expressing cells (C). FISH mapping (bottom) was done as detailed in Materials and Methods. Shown are the FISH signals on the chromosome (left) and the same mitotic figure stained with DAPI to identify chromosome 3 (right).

Biochemical characterization of hSAG. (i) SAG binds to metal ions. We first examined the biochemical activity of SAG. hSAG protein was expressed in bacteria and purified to near homogeneity (submitted for publication). Since SAG is a cysteine-rich protein containing a zinc RING finger domain, it has the potential to bind with metal ions. We used ESI-MS (17) to measure the potential zinc ion binding of SAG by comparing the molecular mass of SAG under denaturing and nondenaturing solution conditions (37, 72). Prior to ESI-MS analysis, urea-denatured pure SAG protein was dialyzed against 50 µM ZnSO₄ for 3 days with three changes of dialysis buffer. Higher concentrations of zinc solution were found to induce precipitation of the protein (data not shown). Under a denaturing acidic solution (pH 2.5 and high organic concentration) where the protein is not expected to retain metal binding characteristics even in the presence of zinc, the molecular mass of SAG was determined to be 12,550 Da, in close agreement with the expected mass for the apoprotein (12,552 Da) (Fig. 5A). ESI-MS analysis of the SAG protein in a nondenaturing aqueous solution (pH 7) resulted in higher masses of 12,733 and 12,800 Da (Fig. 5B), consistent for the holoprotein binding three and four zinc metal ions, respectively.

View larger version (18K): [in this window] [in a new window]   FIG. 5.   SAG is a zinc binding protein. Shown are electrospray mass spectra of SAG protein in denaturing solvent (80:15:5 [vol/vol/vol] acetonitrile-water-acetic acid [pH 2.5]) (A) and nondenaturing solution (10 mM ammonium bicarbonate, 1 mM DTT [pH 7]) (B). The expected molecular mass of the apoprotein is 12,552 Da, with the first methionine deleted during expression and purification in bacteria. Assay conditions are detailed in Materials and Methods.

(ii) SAG prevents LDL oxidation induced by copper ion or a free radical generator. Since SAG binds to metal ions, we reasoned that SAG may prevent metal ion-induced oxidation of macromolecules. Copper-induced oxidation of low-density lipoprotein (LDL) was used as a testing model. As shown in Fig. 6, copper-induced LDL oxidation, as measured by the formation of TBARS, is slightly enhanced by SAG at low concentrations. At higher SAG concentrations, however, a dose-dependent inhibition (up to 90%) of LDL oxidation is observed (Fig. 6A). Inhibition is heat resistant since heat-treated (60°C for 15 min) SAG retains the activity (Fig. 6B), suggesting that enzymatic activity is not involved. Inhibitory activity is, however, completely or partially abolished by pretreatment of SAG with the alkylating reagent N-ethylmaleimide or p-hydroxymercuric benzoate, respectively (Fig. 6B). The results indicate that free SH groups in the SAG molecule are the major contributors to this activity. Furthermore, MT, a small metal ion binding protein consisting of 20 cysteine residues out of 61 amino acids (44), shows an inhibitory curve similar to that for SAG (Fig. 6C). Glutathione, an additional cysteine-containing peptide, showed 25% inhibition at a concentration of 100 µM (data not shown). Inhibition of copper-induced LDL oxidation is, however, not observed in other known antioxidant enzymes such as superoxide dismutase or catalase or other proteins such as bovine serum albumin or cytochrome c (submitted for publication). These results clearly showed that by binding and chelating copper ion through its free SH groups, SAG prevents copper-initiated free radical reactions leading to lipid peroxidation. Superoxide or hydrogen peroxide appears not to be involved in the process. To test whether protection of SAG against LDL oxidation is mediated solely through copper binding, we initiated LDL oxidation by AAPH, a free radical generator (21). In this metal ion-free system, SAG also inhibits LDL oxidation (up to 85%) at a concentration of 59 µM (750 µg/ml) (Fig. 6D). Thus, by metal binding and free radical scavenging, SAG acts as a protector against lipid peroxidation.

View larger version (32K): [in this window] [in a new window]   FIG. 6.   SAG inhibits LDL oxidation induced by copper or AAPH. (A) Dose-dependent inhibition of copper-induced LDL oxidation by SAG. Different amounts of SAG protein were incubated with LDL in the presence of CuSO4 (10 µM) for 10 min. Oxidation of LDL was measured by the formation of TBARS as detailed in Materials and Methods. (B) Abrogation of SAG protection by pretreatment with alkylating reagents but not by heat. SAG (750 µg/ml, 59 µM) was preincubated for 10 min with alkylating reagents N-ethylmaleimide (NEM; 50 mM) and p-hydroxymercuric benzoate (PHMB; 1 µmol/mg of SAG protein) or preheated (60°C for 15 min) before being subjected to LDL oxidation assay. Cont, control. (C) Dose-dependent inhibition of copper-induced LDL oxidation by MT. Different amounts of MT (from rabbit liver or horse kidney) were incubated with LDL in the presence of CuSO4 (10 µM) for 10 min. Oxidation of LDL was measured by the formation of TBARS. (D) Inhibition of LDL oxidation induced by AAPH. SAG (750 µg/ml, 59 µM) was incubated with LDL in the presence of AAPH (5 mM) for 10 min. Formation of TBARS was measured as a index of LDL oxidation.

SAG protects cells from apoptosis induced by redox-sensitive reagents. Apoptosis has been widely implicated in human diseases such as cancers and neurodegenerative disorders (66). Since (i) SAG is inducible concurrent with the OP-induced apoptosis pathway, (ii) bacterially expressed SAG binds to metal ions, zinc, and copper, which also mediate apoptosis (19, 67), and (iii) bacterially expressed hSAG prevents copper or ROS-induced lipid peroxidation, we examined the potential role of SAG in redox-induced apoptosis. Several stable hSAG transfectants were generated in DLD-1 human colon carcinoma cells. Two transfectants (D12-1 and D12-8) showed high expression of SAG mRNA and protein compared to the neo controls (D1-3 and D1-6) (Fig. 7A). Sensitivity to apoptosis induced by OP as well as metal ions (zinc and copper) was examined by morphological appearance and DNA fragmentation. Treatment of cells with 150 µM OP or 125 µM ZnSO₄ for 24 h induced marked cell shrinkage and detachment in the neo control cells. These morphological signs of apoptosis were significantly reduced in SAG-expressing cells (data not shown). In the same treated cells, a DNA fragmentation ladder is clearly seen in two neo control lines (Fig. 7B [lanes 1 and 3] and C [lanes 1 and 2]) but significantly reduced in two SAG-overexpressing transfectants (Fig. 7B [lanes 2 and 4] and C [lanes 3 and 4]). In the case of copper sensitivity (up to 750 µM), expression of SAG does not offer significant protection when judged by morphological signs of apoptosis (not shown). Higher doses induce apoptosis in both lines (data not shown). These results indicated that when overexpressed, SAG protected DLD-1 colon carcinoma cells from apoptosis induced by OP and zinc.

View larger version (57K): [in this window] [in a new window]   FIG. 7.   Overexpression of hSAG protects DLD-1 colon carcinoma cells from apoptosis induced by OP and zinc. (A) Selection of SAG-expressing stable clones. DLD-1 cells were transfected with the neo control pcDNA3 or hSAG expression plasmid pcDNA-SAG. After G418 selection, resistant colonies were ring cloned and subjected to detection of exogenous SAG expression. (a) Expression of SAG mRNA. Total RNA was isolated and subjected to Northern analysis. Vector controls, D1-3 and D1-6; hSAG transfectants, D12-1 and D12-8. (b) 28S and 18S rRNAs for loading controls. (c) Expression of SAG protein in transfectants. The vector control lines and hSAG transfectants were subjected to immunoprecipitation. Shown is SAG protein expression in the neo controls (D1-3 and D1-6) and hSAG transfectants (D12-1 and D12-8). (B and C) hSAG overexpression protects cells from DNA fragmentation induced by OP or zinc. hSAG transfectants (D12-1 and D12-8), along with the vector control cells (D1-3 and D1-6), were seeded at 3.0 × 106 to 3.5 × 106 per 100-mm-diameter dish and exposed after 16 to 24 h to OP (150 µM; B) or zinc sulfate (125 µM; C) for 24 h. Both detached and attached cells in 2- by 100-mm dishes were harvested and subjected to DNA fragmentation assay. The 100-bp size marker is shown in the leftmost lane.

View larger version (137K): [in this window] [in a new window]   FIG. 8.   Overexpression of hSAG protects SY5Y neuroblastoma cells from apoptosis induced by copper and zinc. (A) Selection of SAG-expressing stable clones. SY5Y cells were transfected with the neo control pcDNA3 or hSAG expression plasmid pcDNA-Flag-SAG. After G418 selection, resistant cell lines were ring cloned and subjected to Western blot analysis for exogenous SAG expression with anti-Flag antibody. (B) Protection of metal-induced apoptosis as shown by morphological appearance. SAG-expressing cells (SYW20 [b, d, and f]) and neo control cells (SYV3 [a, c, and e]) were untreated (a and b) or exposed to CuSO4 (1.25 mM [c and d]) or ZnSO4 (200 µM [e and f]) for 16 h. Morphology was visualized at a magnification of ×200. (C) Protection of metal-induced apoptosis as shown by TUNEL assay. Cells were exposed to metal ions for 16 h as described above (without treatment [a and b]; with 1.25 mM CuSO4 [c and d]; with 200 µM ZnSO4 [e and f]), then subjected to TUNEL assay as detailed in Materials and Methods, and analyzed under a fluorescence microscope with a blue filter. Magnification, ×200.

Metal ions induce cytochrome c release and caspase activation that can be inhibited or delayed by SAG expression. Although metal ions have been shown to induce apoptosis through ROS generation, the mechanism of action underlying their effect is not clear. Since cytochrome c release from mitochondria and subsequent caspase activation are the key events in apoptosis (30, 35, 36, 43, 74), levels of cytochrome c released into the cytoplasm and potential activation of caspase were examined following treatment with metal ions. As shown in Fig. 9A (top), there is a low basal level of cytoplasmic cytochrome c in DLD-1 cells. Treatment of cells with 140 µM ZnSO₄ induces the release of cytochrome c into the cytoplasm. Densitometric analysis shows a 2.7- or 2.26-fold increase of cytoplasmic cytochrome c in the vector control D1-6 cells at 16 or 24 h posttreatment, respectively. In SAG-expressing D12-1 cells, an increase of cytoplasmic cytochrome c does not occur until 24 h and reaches a lower level (0.87- or 1.5-fold increase over the control at 16 or 24 h, respectively). Likewise, activation of caspase 7, inferred from the disappearance of the proenzyme form, is seen in a time-dependent manner after zinc treatment (bottom panel). More extensive activation (disappearance of the band) is observed in the vector-transfected cells (0.6- or 0.47-fold below that in untreated control at 16 or 24 h, respectively) than in SAG-expressing cells (1- or 0.68-fold lower than the control value at 16 h or 24 h, respectively). A similar result was obtained with caspase 3 activation (data not shown).

View larger version (78K): [in this window] [in a new window]   FIG. 9.   Overexpression of SAG inhibits or delays metal ion-induced cytochrome c release and caspase activation. (A) SAG-expressing DLD cells (D12-1) and neo control cells (D1-6) were subjected to ZnSO4 (140 µM) treatment for the indicated periods of time. The cytoplasmic fraction was extracted and subjected to Western blot analysis with antibodies against cytochrome c (top) and caspase 7 (bottom). (B) Human 293 cells were transiently transfected with hSAG expression plasmid or the neo control. Twenty-four hours posttransfection, cells were treated with CuSO4 (2 mM) for the indicated periods of time and subjected to Western blot analysis with antibodies against cytochrome c (top) and caspase 7 (bottom). Densitometric quantitation was performed in a densitometer. The band density from the untreated control was arbitrarily chosen as 1.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

SAG is a novel member of the zinc RING finger family of proteins. Using the DD technique, we identified an OP-inducible gene, SAG, that encodes a novel protein containing zinc RING finger domain. Zinc RING finger genes represent is a new and growing family of genes (20, 52) whose members contain C₃HC₄ or C₃H₂C₃ (a cysteine-to-histidine change) motif and are involved biochemically in DNA binding, RNA binding, and protein-protein interactions (8, 9, 16, 38). Biologically, the zinc RING finger proteins are involved in many processes, including oncogenesis, signal transduction, and development, among others (52). Some RING finger proteins, including the baculovirus protein p35 and mammalian homologs of baculovirus inhibitors of apoptosis, are known to inhibit apoptosis (13, 63, 68). SAG, described here, consists of 113 amino acids with 12 cysteine residues. It is the smallest of the RING finger proteins, with 50% of the polypeptide consisting of a RING domain. The protective activity against redox-induced apoptosis renders SAG the first antioxidant molecule among RING finger family members.

SAG is evolutionarily conserved and expressed highly in energy-consuming organs. SAG is evolutionarily conserved, with 55 or 70% identity, respectively, between human and yeast or human and C. elegans sequences. High sequence homology often suggests a functional conservation, as seen in many other proteins, including the apoptosis-related proteins Bcl-2, Apaf-3/caspase 9, and Apaf-1 and their C. elegans homologs, ced-9, ced-3, and ced-4, respectively (35, 69, 75). Indeed, hSAG and ySAG (yeast homolog of hSAG) have similar functions: yeast lethality induced by disruption of ySAG can be fully complemented by hSAG (unpublished data). Thus, hSAG could be a growth-essential gene by functioning as an antioxidant molecule to protect cells against ROS-induced death or by promoting cell growth through other mechanisms (unpublished data).

SAG is an antioxidant molecule that inhibits apoptosis in multiple human tumor lines. In an attempt to isolate genes that mediate OP-induced apoptosis (58), we used the DD technique and isolated two OP-inducible genes. The first gene encodes GSS, the enzyme involved in the last step of glutathione synthesis (57), and the second is SAG. Both seem to be involved in antioxidant pathways that counteract rather than potentiate OP-induced apoptosis (reference 57 and this report). Induction of GSS and SAG, therefore, appears to be a cellular defense response against OP-induced redox disturbance. When this disturbance is dominant and overcomes the protective response, cells eventually undergo apoptosis. In addition to being inducible by the redox agent OP in mouse cell lines, SAG was also induced in mouse brain after ischemia damage (unpublished observation). Thus, SAG appears to be a stress-responsive gene in cells that acts as a protector against ROS-induced damage.